Find flow-based communities in complex networks

This release makes Infomap's results more correct. Hierarchical solutions get about 0.2–0.5% shorter codelengths, because modules that do not pay for their codebook are now removed. If you rerun an analysis made with an earlier version, expect the partition to change even with the same seed. If you used --entropy-corrected, --regularized or --meta-data, rerun: fixes to these options change the codelength and the modules found. The entropy correction, for example, was counted in nats instead of bits. To tell an intended change from a regression, check the new list of result changes by release and flag. Output files now record the seed and a fingerprint of the input, so a result can be traced to the run that made it (changelog).

Network Navigator is rebuilt on d3gl and now shows raw or clustered networks with millions of links in the browser. Open an Infomap .ftree result, or load a raw network and cluster it with Infomap without leaving the page. Choose a plain or nested layout. The level of detail, based on the module hierarchy or on spatial proximity, expands continuously as you zoom in. State networks can be shown with physical nodes, state nodes, or both. Export the view as SVG or PNG, or the result as .ftree (GitHub).

C3 generates distinct, deterministic categorical colors for maps and charts. Version 2 places hues on the OKLCH hue circle or along any d3 color scheme, and adds tones, seed colors, and nested colors for hierarchies, with each child varying within its parent's range. It now comes in TypeScript (npm install @mapequation/c3) and Python (pip install c3-colors) with the same functions and the same colors, and has an interactive explorer (changelog).

infomap.run() is now the canonical, fully typed entry point for community detection — common options live on the signature, and results come back as rich Result objects with summary(), to_series(), and one-call to_networkx() and to_igraph() conversions, plus a summary card that renders inline in notebooks. Also new: bundled example networks in infomap.datasets, named state nodes across the Python and R APIs, the infomap-network v1.0 JSON input format, and exposed flow-iteration and tolerance controls. Options are now validated with guided error messages, and engine logging routes through Python's standard logging module (Python docs, changelog).

The map equation tutorial interactively explains the mechanics of the Map equation. In his master's thesis work, Henrik Larsson did a major update of the whole app and added missing sections explaining the hierarchical map equation and regularization for incomplete data.